AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splats

Chen Ziwen, Hao Tan, Kai Zhang, Sai Bi, Fujun Luan, Yicong Hong, Li Fuxin, Zexiang Xu

ICCV 2025 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting · 63 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace optimization-based 3D GS with feed-forward Long-LRM at 32 viewsLong-LRM · DL3DV-140 (32 views)reconstruction time: 13min -> 1secImproved
add 10 post-prediction optimization stepsLong-LRM · DL3DV-140 (32 views)PSNR: 24.10 -> 25.60Improved
Long-LRM versus optimization-based methods on perceptual qualityLong-LRM · DL3DV-140 (32 views)LPIPS: 0.205 -> 0.254Hurt
replace all-Transformer with hybrid Mamba2+Transformer blocks (ablation)Long-LRM · 32 views 256 resPSNR: 24.28 -> 26.82Improved
use hybrid instead of all-Transformer at 4 views (ablation)Long-LRM · 4 views 256 resPSNR: 21.13 -> 21.58Improved
add token merging with token dimensionality 256 to 1024 (ablation)Long-LRM hybrid · 32 views 512 resiteration time (sec): 11.5 -> 4.0Improved
add token merging with token dimensionality 256 to 1024 (ablation)Long-LRM hybrid · 32 views 512 resPSNR: 28.16 -> 27.46Hurt
add opacity regularization loss (ablation)Long-LRM · 4 views 256 res% Gaussians with opacity>0.001: 99.2 -> 68.3Improved
add depth supervision from DepthAnything (ablation)Long-LRM · 4 views 256 resPSNR: 20.96 -> 21.25Improved
add opacity and depth losses on two-view RealEstate10K (ablation)Long-LRM · RealEstate10K (2 views)PSNR: 28.54 -> 28.44No clear effect
concentrate transformer blocks (3T21M or 21M3T) instead of evenly distributing themLong-LRM hybrid · 4 views 256 restraining stabilityHurt
fine-tune Long-LRM to 2D GSLong-LRM · ScanNetv2 zero-shotPSNR: 19.25 -> 23.27Improved

Later papers that compared with this method

7 later accepted papers report Long-LRM as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy